Influence of somatotropin and IGF gene expression C. K. WOLVERTON,
on lipid metabolism in porcine adipose tissue
M. J. AZAIN,
J. Y. DUFFY,
M. E. WHITE,
AND T. G. RAMSAY
Department of Animal Science, Ohio State University, Columbus, Ohio 43210; and Department of Animal and Dairy Science, University of Georgia, Athens, Georgia 30602 Wolverton, C. K., M. J. Azain, J. Y. Duffy, M. E. White, and T. G. Ramsay. Influence of somatotropin on lipid metabolism and IGF gene expression in porcine adipose tissue. Am. J. Physiol. 263 (Endocrinol. Metab. 26): E637-E645, 1992. -The present study was designed to evaluate the effects of porcine somatotropin (PST) treatment (2 mg/day) and dietary fat (10%) separately and in combination on the metabolic activity of subcutaneous adipose tissue, serum adipogenic activity, and insulin-like growth factor (IGF) gene expression within adipose tissue from growing 5- to 6-mo-old barrows. This study attempted to determine how these factors might contribute to the reported changes in adiposity of treated swine. Biopsies of adipose tissue were collected after 28 days of treatment following anesthesia with thiopental sodium (15 mg/kg iv). Somatotropin inhibited in vitro glucose oxidation and lipogenesis in adipose tissue but did not affect fatty acid esterification. Adipogenic activity of serum was not altered by pST treatment. Subcutaneous adipose tissue contained mRNA for IGF-I and -11, and pST administration increased the abundance of IGF-I mRNA. Dietary fat had no effect on these variables. Thus somatotropin reduces glucose metabolism in porcine subcutaneous adipose tissue. Preadipocyte proliferation and differentiation are not affected by somatotropin through its actions on systemic factors. Dietary fat provides no additional benefit in combination with pST administration to affect accretion of adipose tissue in growing swine. growth hormone; swine; lipogenesis;
somatomedin; insulin-like adipogenic activity; fat cell
growth
factor;
OF porcine somatotropin (PST) to growing swine causes a decrease in adipose tissue lipid accretion (8, 13, 14). Two major mechanisms that can affect the accumulation of adipose tissue are alterations in adipocyte metabolism or changes in preadipocyte proliferation and differentiation. The mechanisms by which pST influences the deposition of adipose tissue are unclear. Experiments with isolated pig adipocytes indicate that acute exposure to pST has lipogenic effects while chronic exposure to somatotropin has antilipogenie effects (22, 38). Somatotropin either has no effect or inhibits differentiation, while decreasing the number and size of fat cell clusters in cultures derived from neonatal porcine adipose tissue (19). Somatotropin also antagonizes lipid accumulation by adipocytes in rat and pig stromal-vascular cell cultures (31, 34), as well as antagonizing insulin-stimulated adipogenesis (19). Many of PST’S effects on growth are mediated by insulin-like growth factor I (IGF-I). Similar to somatotropin, IGF-I may exert lipogenic activity within adipocytes in vitro (16). IGF-I has been demonstrated to influence the proliferation and differentiation of porcine preadipocytes as well as several established cell lines (33, 42). Recently, several IGF-I mRNA transcripts from Ob1771 cells and isolated rat and pig adipocytes have been identified (11, 17). These IGF mRNA species appear to be regulated by somatotropin in vitro. Thus ADMINISTRATION
0193-1849/92
$2.00
Copyright
IGF-I may be a paracrine factor produced within the adipose tissue which can promote the proliferation and differentiation of the preadipocyte or stimulate the metabolism of adipocytes within that tissue. This effect by IGF-I on the cellularity and metabolism of porcine adipose tissue is in direct contrast to the previously reported actions of pST on adipose tissue; the administration of pST increases serum IGF-I concentration while decreasing deposition of adipose tissue and carcass lipid in vivo (13, 14). Dietary fat inhibits fatty acid synthesis in adipose tissue and exhibits an overall glucose-sparing effect (1). Administration of pST and feeding a high-fat diet improved average daily gain in young growing swine (5). Feeding swine a high-fat diet and giving PST had an additive effect on the inhibition of glucose utilization by porcine adipose tissue (3). These data suggest that supplementing dietary fat to PST-administered pigs increases the pST effects on porcine adipose tissue. IGF-I is sensitive to nutritional alterations. Serum concentration of IGF-I decreases with fasting and protein restriction and returns to normal with refeeding or increasing dietary protein (7, 24). The impact of dietary fat on pST induction of IGF-I synthesis or IGF-I mRNA is unknown. Further information concerning the mechanisms of pST action and those mechanisms relating pST to IGF expression need to be accumulated both at the cellular and molecular level to better clarify the roles of pST and IGF in regulation of adipose tissue development and metabolism in the pig. The present study was performed to investigate the actions of pST and dietary fat on the metabolism of adipose tissue and to determine if pST alters the adipogenic activity of serum from treated pigs. Second, this study was performed to determine if pST or dietary fat can alter the expression of IGF-I and insulinlike growth factor II (IGF-II) in porcine adipose tissue. METHODS
Sixteen individually penned, castrated male pigs (Yorkshire Duroc) at 5-6 mo of age were utilized in this study. All live animal procedures were conducted at the University of Georgia Swine Center in January-February, 1989. Animals were divided into four treatment groups with four pigs in each group. Treatment groups were 1) control diet, vehicle injected; 2) control diet, 2 mg PST/day; 3) fat diet (10% supplemental fat), vehicle injected; and 4) fat diet, 2 mg PST/day. Treatments were initiated at an average body weight of 83 kg and continued for 4 wk. Diets are described in Table 1. Control and fat diets were formulated to meet or exceed the National Research Council requirements for the pig and the PST-treated pig, and they were offered ad libitum. The composition of the diets has been described previously (5). Briefly, the control diet contained (calculated) 14.0% crude protein, 0.96% total lysine, and 3,210 kcal metabolizable energy/kg. The fat diet contained 10% poultry fat x
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpendo at Washington Univ (128.252.067.066) on February 12, 2019.
E637
E638
SOMATOTROPIN
AND
Table 1. Diet composition Item
Pretreatment
Ingredient Corn Soybean meal Lysine HCl Poultry fat Dicalcium phosphate Limestone Salt Trace mineral mix Selenium mix Vitamin mix Calculated analysis Crude protein, % Lysine, % Malic enzyme, kcal/kg Calorie/protein, kcal/g
Control
Fat
81.60 15.80
82.90 14.00 0.40
1.09 0.66 0.40 0.15 0.10 0.25
1.14 0.66 0.40 0.15 0.10 0.25
65.40 21.60 0.40 10.00 1.10 0.61 0.40 0.15 0.10 0.25
14.40 0.69 3,228 22.40
14.00 0.96 3,210 22.90
16.30 1.16 3,678 22.60
Minerals provided are as follows (in mg/kg diet): 75 Mn, 60 Zn, 25 Fe, 3 Cu, and 1 I. Selenium in the diet was 0.3 mg/kg diet. Vitamin mix contained the following (per kg diet): A, 4,961 IU; DIl, 882 IU; E, 33 IU; K, 3.3 mg; riboflavin, 4.4 mg; pantothenic acid, 22 mg; niacin, 27.6 mg; choline, 110 mg; B12, 22 pg; biotin, 55 pg.
and was formulated so that the calorie-to-protein (and calorieto-lysine) ratio was similar to the control diet. This diet was calculated to contain 16.3% crude protein, 1.16% total lysine, and 3,670 kcal metabolizable energy/kg. Recombinantly derived pST used in the study was provided by Monsanto (St. Louis, MO). The pST (2 mg/day) was administered daily between 0800-1000 for 28 days to pigs in groups 2 and 4. Animals in groups 1 and 3 received an injection of vehicle (25 mM sodium bicarbonate buffer, pH 9.5). One animal in group 2 broke a leg before the end of the study and was killed. Data are not presented for this animal. Blood was collected from the jugular vena cava 3 h after the last injection of pST or vehicle on the 28th day of the study and before adipose tissue biopsy. Serum was processed for each animal and frozen for later analysis of adipogenic activity, serum growth hormone (PST), and IGF-I concentrations. Serum IGF-I levels were determined after acid ethanol extraction by using a heterologous radioimmunoassay (RIA) kit (Nichols Institute Diagnostics, San Juan Capistrano, CA), and serum growth hormone levels were determined by a homologous RIA (Monsanto, Chesterfield, MO). Sera were also tested on stromal-vascular cell cultures from subcutaneous adipose tissue of neonatal swine to determine if any alteration of the serum profile would influence the growth and metabolism of porcine preadipocytes in culture. In vitro metabolism. Subcutaneous adipose tissue was obtained at biopsy. Briefly, pigs were restrained 3 h after injection of somatotropin or vehicle on the 28th day of the study, and a jugular blood sample was obtained. Subsequently, thiopental sodium (15 mg/kg) was injected intravenously. Tissue samples were obtained from over the shoulders after the site was shaved and cleaned. Tissue plugs were placed in saline (tissue slice incubations) or frozen immediately in liquid nitrogen (cellularity, enzyme activity, mRNA analysis). Tissue slices of - 100 mg were prepared from the inner layer of adipose tissue using a Stadie-Riggs microtome. Tissues were incubated for 2 h in 25ml flasks in 2 ml of Krebs-Henseleit buffer with 2% bovine serum albumin and containing 5 mM glucose (0.5 &i [U-14C]glucose/ml medium) or 1.0 mM palmitate (1 .O &i [ 1-14C]palmitate/ml medium). Triplicate slices were prepared for each animal and each substrate condition. Linearity of metabolism with time through 3 h of incubation was demonstrated (data not shown). Incubations were termi-
ADIPOSE
TISSUE
nated by the injection of 1 N H,SO, into the medium. Incorporation of [14C]glucose into COZ, total lipids, and the fatty acid portion of total lipids, as well as the esterification of palmitate, was determined as previously described (4). Negligible amounts of palmitate label were recovered as CO2 or ketones and thus are not reported in the present study. Lipogenic enzyme assays. Frozen adipose tissue slices of 500 mg were homogenized in 1.5 ml of ice-cold homogenizing buffer [0.25 M sucrose, 0.001 M EDTA, 0.005 M tris(hydroxymethyl)aminomethane (Tris) base, and 0.001 M dithiothreitol, pH 7.41. Homogenates were centrifuged at 28,000 g for 20 min at 4°C. Supernatant was collected and centrifuged for 20 min at 4°C at 1,200 g to remove residual lipid droplets. Resulting supernatants were kept on ice and were used for the enzyme assays. Enzymes assayed were the following: citrate lyase (CL), glucose-6-phosphate dehydrogenase (G-6-PDH), 6-phosphogluconate dehydrogenase (6-PGDH), sn-glycerol-3-phosphate dehydrogenase (GPDH), malic enzyme (ME), fatty-acid synthase (FAS), and lipoprotein lipase (LPL). CL activity was assessed according to Cottam and Srere (8). G-6-PDH/G-PGDH activities were determined by the double substrate assay of Glock and McLean (18). GPDH was assayed according to the procedures of Wise and Green (41) as modified by Ramsay et al. (30). ME activity was assayed according to the procedures of Ochoa (29). Changes in absorption at 340 nm were measured in a Gilford Response spectrophotometer at 25°C. Enzyme activities were expressed as nanomoles of substrate utilized per minute. FAS was assayed by the procedures of Nepokroeff et al. (26); one unit of enzyme activity was equal to 1 nmol [I-‘“Clacetyl CoA incorporated/lo min. LPL activity was assessed according to the procedures of Nilsson-Ehle and Schotz (27); one unit of activity was equivalent to 1 nmol [“HIoleic acid released/30 min. All assays were linear for sample concentration and time. Data are expressed as units of activity per 1 x 10” cells. Adipocyte number was determined by using osmium tetroxide fixation and electronic counting according to the procedures of Etherton et al. (12). Cell culture. Sera were tested on stromal-vascular cell cultures prepared from the subcutaneous adipose tissue of neonatal swine. Adipose tissue was excised between the first and second thoracic vertebrae just lateral to the dorsal midline by sterile dissection. Tissues were minced, and enzymatic digestion, filtration, and centrifugation were utilized to isolate the stromalvascular cells (containing a population of preadipocytes) according to the methods of Ramsay et al. (30). Plating medium consisted of medium 199 with Earle’s salts, 5 mM glucose, 40 mg/l gentamicin sulfate, 50 mg/l cephalothin, 2 mg/l Fungizone, and 10% fetal bovine serum. Test media contained the same components as the plating medium except for serum source. Stromal-vascular cells were seeded in 25-cm2 flasks and 35-mm tissue culture dishes at a density of 1 x lo4 cells/cm2. Cells were cultured at 37°C in a humidified 5% CO, atmosphere. Plating medium was replaced with growth medium containing 2.5% porcine serum (Hyclone, Logan, UT) in medium 199 after 24 h of incubation. Growth medium was changed on days 1 and 3 of culture. Medium was changed to test media containing 2.5% test sera in medium 199 on day 5 of culture (confluency). Test media were changed every 3 days of culture. Cells were harvested on day 14 of culture and analyzed for activities of GPDH and LPL and for DNA content (15). Additional cells were plated in 25-cm2 flasks to evaluate glucose metabolism by the cells following differentiation in response to the test sera, according to previously published procedures (32). Briefly, cells were incubated with 2.5% porcine serum (Hyclone) in medium 199 from day 1 to day 5 of culture. Medium was changed to test media containing 2.5% test sera in medium 199 on day 5 of culture. Test media were changed every 3 days, and glucose metabolism was evaluated on day 14 of
Downloaded from www.physiology.org/journal/ajpendo at Washington Univ (128.252.067.066) on February 12, 2019.
SOMATOTROPIN
AND ADIPOSE
Cellswere washedfor 1 h with Earle’sbalancedsaltsto remove any medium contaminants and switched to incubation medium containing medium 199, 25 mM N-2-hydroxyethylpiperazine-N’ -2-ethanesulfonicacid (HEPES), 10 mM glucose,1 &i/ml [U-14C]glucose,and 2% bovine serumalbumin for 4 h. CO, was collected in plastic wells containing hyamine hydroxide, and the incorporated radioactivity was measuredby scintillation counting. Flask contents (porcine adipocytes) were analyzed for [ 14C]glucose incorporation into COe,total lipids, and fatty acids according to the procedure of DeCingolani (10). Isolation of total RNA. Tissue sampleswere pooled from animals in each treatment group. Total RNA was extracted and isolated using the guanidine isothiocyanate-cesium chloride method (6). Pooled tissues (5 g) were homogenized in 4 M guanidine isothiocyanate and centrifuged through a cushion of 5.7 M cesium chloride for 22 h at 20°C. RNA was phenolchloroform extracted and ethanol precipitated. Total RNA was quantified spectrophotometrically at 260 nm in a Gilford Responsespectrophotometer,then electrophoresedthrough a 1.2% denaturing agarosegel and stained with ethidium bromide to determine integrity of the RNA. All RNAs were judged to be intact by the presenceof 28s and 18s ribosomalbandsat a mass ratio of -2:l. Dot blot. Total RNA wasdenatured in 50% deionized formamide, 6% formaldehyde, and 20 mM Tris, pH 7.0. Various concentrations of total RNA were transferred to a charged nylon membrane (Immobilon, Millipore) using a dot-blot vacuum manifold (BRL, Gaithersburg, MD). Yeast tRNA was usedas the negative control and did not hybridize with the probes. Levels of mRNA on dot blots were estimated by laser densitometric scanning (LKB Ultrascan XL scanninglaserdensitometer) and by direct comparison to known concentrations of a cRNA standard curve present on the sameblot, according to procedurespreviously described(21). Northern analysis. Total RNA was isolated, and 20 pg were denaturedat 65°C for 5 min with 50% deionized formamide, 2.2 M formaldehyde, 24 mM HEPES, 6 mM sodiumacetate trihydrate, and 1.2 mM EDTA. Total RNA was separatedby electrophoresisin a 1.2% agarose-2.2M formaldehyde geland transferred to a nylon membrane(Immobilon, Millipore) by capillary elution transfer according to the procedure of Maniatis et al. (23). RNA filters were immobilized by baking in a vacuum oven for 2 h at 80°C. Ethidium bromide staining of ribosomalRNA was usedto monitor RNA integrity and equivalent loading of samples.Quantitative and equivalent transfer was verified by ethidium bromide staining and ultraviolet analysis of the gel following transfer. Nick trandation. Probeswere labeledby nick translation with specific activity of 1.5-2.0 x lo8 countsmin-l .pg DNA? Membranes for both dot-blot and Northern analysis were prehybridized in 5 ml of Rapid Hybridization Buffer (Amersham), 100 bg/ml yeast tRNA, and 1% Triton X-100 at 70°C for 30 min. Hybridization wasperformed at 65°C for 18 h. Membranes were rinsed at 65°C with 2~ SSC, 0.1% sodiumdodecyl sulfate (SDS) twice for 15 min, two times with IX SSC, 0.1% SDS, and twice in 0.1x WC, 0.1% SDS for 15 min each. All membranes wereautoradiographedusing Kodak XAR X-ray film at -70°C. Probes.The 580-bpporcine IGF-I cDNA wasa gift from Dr. Frank Simmen, and the rat 720-bp IGF-II cDNA was kindly provided by Dr. M. Rechler. The 720-bp rat IGF-I cDNA used for the cRNA transcripts was provided by Dr. P. Rotwein. Statistical analysis. Data were analyzed by two-way analysis of variance using SAS (35). The model includedthe main effects of diet, pST treatment, and the interaction of diet and treatment. Differences between means were tested using Fisher’s least significant difference when effects were significant (P < 0.05). culture.
E639
TISSUE
RESULTS
There was no significant effect of diet or somatotropin on rate of gain of animals in this study. The lack of effect of somatotropin on gain is in contrast to other studies (5, 13, 14) and can likely be accounted for by the small number of animals used in the present study. However, PST treatment did cause a reduction in feed consumption (-27%, P < 0.0001) and an improvement in feed conversion (feed-to-gain ratio: vehicle, 3.74; PST, 2.78; P < O.OOl), in agreement with previous work (5, 13, 14). Similarly, dietary fat resulted in a reduced feed intake (-8%; P < 0.05) and an improved feed-to-gain ratio (control, 3.67; fat 2.85; P c 0.0001). The effects of dietary fat were accounted for by differences in energy density of the diet. Thus, on a caloric basis, intake and feed conversion of animals fed fat were similar to those of animals fed the control diet. These results are similar to those reported previously (5). The effects of pST on serum growth hormone and IGF-I concentration are shown in Fig. 1. Treatment with exogenous pST elevated serum growth hormone and IGF-I concentrations in comparison with treatment with vehicle, independent of diet. Analysis of biopsies of subcutaneous adipose tissue which were utilized for tissue incubations and enzyme assays indicated that pST treatment and dietary fat affected the cellularity of the biopsies (Fig. 2). PST treatment resulted in more adipocytes per gram tissue than in tissue from control animals by determination of average cell number per gram tissue by osmium tetroxide fixation and electronic counting. Feeding of a high-fat diet resulted in fewer adipocytes per gram of tissue than in animals fed the control diet. Treatment with pST did not overcome the effect of fat feeding on the cellularity of the samples of subcutaneous adipose tissue. Treatment with exogenous pST resulted in an inhibition of glucose metabolism within the adipose tissue of treated swine when the results are expressed per lo6 cells (Fig. 3). Glucose conversion to carbon dioxide, total lipids, and fatty acids was reduced by at least 60%. Feeding of a diet supplemented with fat did not alter this pST response. Somatotropin treatment had minor inhibitory effects 2oflfl400 0
TRTl
TRT2
'
PST
0
~-1
1350
TRT3
Fig. 1. Effect of porcine somatotropin (PST) and dietary fat supplementation on serum PST and insulin-like growth factor I (IGF-I) concentration. Values represent mean of triplicate determinations of pooled samples as described in METHODS. TRT 1,O mg PST/day; TRT 2,2 mg PST/day; TRT 3,0 mg pST/day, 10% added fat; TRT 4,2 mgPST/day, 10% added fat.
Downloaded from www.physiology.org/journal/ajpendo at Washington Univ (128.252.067.066) on February 12, 2019.
E640
SOMATOTROPIN
AND ADIPOSE
1.4 Pooled n
4”
SE
0.62
g 3.5-’ .:.
-/
j= E cn
C. :s >: :ij: ::: j:i :t. j:::; :c: :;:.: j: :‘: j:::: ;$ ::: ;.i $ g :i: ... :.: j:.:. ::: ::: :.: ;g :i: ::: # :‘: :7 j: ;.: j: ;.: F. :_ 5. ;:I ;:.:. :::::
-/
yO.8-’ -/
;0.6-’ -E
-/
z’ 0.4-’ s -,
O-2-’ / 0’
8 “+
p ‘.. ._.
b
l-’
E
Pooled
C
I .2-’ 5 E
TISSUE
I
TRT 1
TRT 2
TRT 3
3- I
5 2.5-’ Q) ‘is E 2- Z
::: j::: ::: j:::: j:::: (..:. :.: :.: ::: :.: ::: :c :.. .:.
on lipid metabolism in porcine adipose tissue
M. J. AZAIN,
J. Y. DUFFY,
M. E. WHITE,
AND T. G. RAMSAY
Department of Animal Science, Ohio State University, Columbus, Ohio 43210; and Department of Animal and Dairy Science, University of Georgia, Athens, Georgia 30602 Wolverton, C. K., M. J. Azain, J. Y. Duffy, M. E. White, and T. G. Ramsay. Influence of somatotropin on lipid metabolism and IGF gene expression in porcine adipose tissue. Am. J. Physiol. 263 (Endocrinol. Metab. 26): E637-E645, 1992. -The present study was designed to evaluate the effects of porcine somatotropin (PST) treatment (2 mg/day) and dietary fat (10%) separately and in combination on the metabolic activity of subcutaneous adipose tissue, serum adipogenic activity, and insulin-like growth factor (IGF) gene expression within adipose tissue from growing 5- to 6-mo-old barrows. This study attempted to determine how these factors might contribute to the reported changes in adiposity of treated swine. Biopsies of adipose tissue were collected after 28 days of treatment following anesthesia with thiopental sodium (15 mg/kg iv). Somatotropin inhibited in vitro glucose oxidation and lipogenesis in adipose tissue but did not affect fatty acid esterification. Adipogenic activity of serum was not altered by pST treatment. Subcutaneous adipose tissue contained mRNA for IGF-I and -11, and pST administration increased the abundance of IGF-I mRNA. Dietary fat had no effect on these variables. Thus somatotropin reduces glucose metabolism in porcine subcutaneous adipose tissue. Preadipocyte proliferation and differentiation are not affected by somatotropin through its actions on systemic factors. Dietary fat provides no additional benefit in combination with pST administration to affect accretion of adipose tissue in growing swine. growth hormone; swine; lipogenesis;
somatomedin; insulin-like adipogenic activity; fat cell
growth
factor;
OF porcine somatotropin (PST) to growing swine causes a decrease in adipose tissue lipid accretion (8, 13, 14). Two major mechanisms that can affect the accumulation of adipose tissue are alterations in adipocyte metabolism or changes in preadipocyte proliferation and differentiation. The mechanisms by which pST influences the deposition of adipose tissue are unclear. Experiments with isolated pig adipocytes indicate that acute exposure to pST has lipogenic effects while chronic exposure to somatotropin has antilipogenie effects (22, 38). Somatotropin either has no effect or inhibits differentiation, while decreasing the number and size of fat cell clusters in cultures derived from neonatal porcine adipose tissue (19). Somatotropin also antagonizes lipid accumulation by adipocytes in rat and pig stromal-vascular cell cultures (31, 34), as well as antagonizing insulin-stimulated adipogenesis (19). Many of PST’S effects on growth are mediated by insulin-like growth factor I (IGF-I). Similar to somatotropin, IGF-I may exert lipogenic activity within adipocytes in vitro (16). IGF-I has been demonstrated to influence the proliferation and differentiation of porcine preadipocytes as well as several established cell lines (33, 42). Recently, several IGF-I mRNA transcripts from Ob1771 cells and isolated rat and pig adipocytes have been identified (11, 17). These IGF mRNA species appear to be regulated by somatotropin in vitro. Thus ADMINISTRATION
0193-1849/92
$2.00
Copyright
IGF-I may be a paracrine factor produced within the adipose tissue which can promote the proliferation and differentiation of the preadipocyte or stimulate the metabolism of adipocytes within that tissue. This effect by IGF-I on the cellularity and metabolism of porcine adipose tissue is in direct contrast to the previously reported actions of pST on adipose tissue; the administration of pST increases serum IGF-I concentration while decreasing deposition of adipose tissue and carcass lipid in vivo (13, 14). Dietary fat inhibits fatty acid synthesis in adipose tissue and exhibits an overall glucose-sparing effect (1). Administration of pST and feeding a high-fat diet improved average daily gain in young growing swine (5). Feeding swine a high-fat diet and giving PST had an additive effect on the inhibition of glucose utilization by porcine adipose tissue (3). These data suggest that supplementing dietary fat to PST-administered pigs increases the pST effects on porcine adipose tissue. IGF-I is sensitive to nutritional alterations. Serum concentration of IGF-I decreases with fasting and protein restriction and returns to normal with refeeding or increasing dietary protein (7, 24). The impact of dietary fat on pST induction of IGF-I synthesis or IGF-I mRNA is unknown. Further information concerning the mechanisms of pST action and those mechanisms relating pST to IGF expression need to be accumulated both at the cellular and molecular level to better clarify the roles of pST and IGF in regulation of adipose tissue development and metabolism in the pig. The present study was performed to investigate the actions of pST and dietary fat on the metabolism of adipose tissue and to determine if pST alters the adipogenic activity of serum from treated pigs. Second, this study was performed to determine if pST or dietary fat can alter the expression of IGF-I and insulinlike growth factor II (IGF-II) in porcine adipose tissue. METHODS
Sixteen individually penned, castrated male pigs (Yorkshire Duroc) at 5-6 mo of age were utilized in this study. All live animal procedures were conducted at the University of Georgia Swine Center in January-February, 1989. Animals were divided into four treatment groups with four pigs in each group. Treatment groups were 1) control diet, vehicle injected; 2) control diet, 2 mg PST/day; 3) fat diet (10% supplemental fat), vehicle injected; and 4) fat diet, 2 mg PST/day. Treatments were initiated at an average body weight of 83 kg and continued for 4 wk. Diets are described in Table 1. Control and fat diets were formulated to meet or exceed the National Research Council requirements for the pig and the PST-treated pig, and they were offered ad libitum. The composition of the diets has been described previously (5). Briefly, the control diet contained (calculated) 14.0% crude protein, 0.96% total lysine, and 3,210 kcal metabolizable energy/kg. The fat diet contained 10% poultry fat x
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpendo at Washington Univ (128.252.067.066) on February 12, 2019.
E637
E638
SOMATOTROPIN
AND
Table 1. Diet composition Item
Pretreatment
Ingredient Corn Soybean meal Lysine HCl Poultry fat Dicalcium phosphate Limestone Salt Trace mineral mix Selenium mix Vitamin mix Calculated analysis Crude protein, % Lysine, % Malic enzyme, kcal/kg Calorie/protein, kcal/g
Control
Fat
81.60 15.80
82.90 14.00 0.40
1.09 0.66 0.40 0.15 0.10 0.25
1.14 0.66 0.40 0.15 0.10 0.25
65.40 21.60 0.40 10.00 1.10 0.61 0.40 0.15 0.10 0.25
14.40 0.69 3,228 22.40
14.00 0.96 3,210 22.90
16.30 1.16 3,678 22.60
Minerals provided are as follows (in mg/kg diet): 75 Mn, 60 Zn, 25 Fe, 3 Cu, and 1 I. Selenium in the diet was 0.3 mg/kg diet. Vitamin mix contained the following (per kg diet): A, 4,961 IU; DIl, 882 IU; E, 33 IU; K, 3.3 mg; riboflavin, 4.4 mg; pantothenic acid, 22 mg; niacin, 27.6 mg; choline, 110 mg; B12, 22 pg; biotin, 55 pg.
and was formulated so that the calorie-to-protein (and calorieto-lysine) ratio was similar to the control diet. This diet was calculated to contain 16.3% crude protein, 1.16% total lysine, and 3,670 kcal metabolizable energy/kg. Recombinantly derived pST used in the study was provided by Monsanto (St. Louis, MO). The pST (2 mg/day) was administered daily between 0800-1000 for 28 days to pigs in groups 2 and 4. Animals in groups 1 and 3 received an injection of vehicle (25 mM sodium bicarbonate buffer, pH 9.5). One animal in group 2 broke a leg before the end of the study and was killed. Data are not presented for this animal. Blood was collected from the jugular vena cava 3 h after the last injection of pST or vehicle on the 28th day of the study and before adipose tissue biopsy. Serum was processed for each animal and frozen for later analysis of adipogenic activity, serum growth hormone (PST), and IGF-I concentrations. Serum IGF-I levels were determined after acid ethanol extraction by using a heterologous radioimmunoassay (RIA) kit (Nichols Institute Diagnostics, San Juan Capistrano, CA), and serum growth hormone levels were determined by a homologous RIA (Monsanto, Chesterfield, MO). Sera were also tested on stromal-vascular cell cultures from subcutaneous adipose tissue of neonatal swine to determine if any alteration of the serum profile would influence the growth and metabolism of porcine preadipocytes in culture. In vitro metabolism. Subcutaneous adipose tissue was obtained at biopsy. Briefly, pigs were restrained 3 h after injection of somatotropin or vehicle on the 28th day of the study, and a jugular blood sample was obtained. Subsequently, thiopental sodium (15 mg/kg) was injected intravenously. Tissue samples were obtained from over the shoulders after the site was shaved and cleaned. Tissue plugs were placed in saline (tissue slice incubations) or frozen immediately in liquid nitrogen (cellularity, enzyme activity, mRNA analysis). Tissue slices of - 100 mg were prepared from the inner layer of adipose tissue using a Stadie-Riggs microtome. Tissues were incubated for 2 h in 25ml flasks in 2 ml of Krebs-Henseleit buffer with 2% bovine serum albumin and containing 5 mM glucose (0.5 &i [U-14C]glucose/ml medium) or 1.0 mM palmitate (1 .O &i [ 1-14C]palmitate/ml medium). Triplicate slices were prepared for each animal and each substrate condition. Linearity of metabolism with time through 3 h of incubation was demonstrated (data not shown). Incubations were termi-
ADIPOSE
TISSUE
nated by the injection of 1 N H,SO, into the medium. Incorporation of [14C]glucose into COZ, total lipids, and the fatty acid portion of total lipids, as well as the esterification of palmitate, was determined as previously described (4). Negligible amounts of palmitate label were recovered as CO2 or ketones and thus are not reported in the present study. Lipogenic enzyme assays. Frozen adipose tissue slices of 500 mg were homogenized in 1.5 ml of ice-cold homogenizing buffer [0.25 M sucrose, 0.001 M EDTA, 0.005 M tris(hydroxymethyl)aminomethane (Tris) base, and 0.001 M dithiothreitol, pH 7.41. Homogenates were centrifuged at 28,000 g for 20 min at 4°C. Supernatant was collected and centrifuged for 20 min at 4°C at 1,200 g to remove residual lipid droplets. Resulting supernatants were kept on ice and were used for the enzyme assays. Enzymes assayed were the following: citrate lyase (CL), glucose-6-phosphate dehydrogenase (G-6-PDH), 6-phosphogluconate dehydrogenase (6-PGDH), sn-glycerol-3-phosphate dehydrogenase (GPDH), malic enzyme (ME), fatty-acid synthase (FAS), and lipoprotein lipase (LPL). CL activity was assessed according to Cottam and Srere (8). G-6-PDH/G-PGDH activities were determined by the double substrate assay of Glock and McLean (18). GPDH was assayed according to the procedures of Wise and Green (41) as modified by Ramsay et al. (30). ME activity was assayed according to the procedures of Ochoa (29). Changes in absorption at 340 nm were measured in a Gilford Response spectrophotometer at 25°C. Enzyme activities were expressed as nanomoles of substrate utilized per minute. FAS was assayed by the procedures of Nepokroeff et al. (26); one unit of enzyme activity was equal to 1 nmol [I-‘“Clacetyl CoA incorporated/lo min. LPL activity was assessed according to the procedures of Nilsson-Ehle and Schotz (27); one unit of activity was equivalent to 1 nmol [“HIoleic acid released/30 min. All assays were linear for sample concentration and time. Data are expressed as units of activity per 1 x 10” cells. Adipocyte number was determined by using osmium tetroxide fixation and electronic counting according to the procedures of Etherton et al. (12). Cell culture. Sera were tested on stromal-vascular cell cultures prepared from the subcutaneous adipose tissue of neonatal swine. Adipose tissue was excised between the first and second thoracic vertebrae just lateral to the dorsal midline by sterile dissection. Tissues were minced, and enzymatic digestion, filtration, and centrifugation were utilized to isolate the stromalvascular cells (containing a population of preadipocytes) according to the methods of Ramsay et al. (30). Plating medium consisted of medium 199 with Earle’s salts, 5 mM glucose, 40 mg/l gentamicin sulfate, 50 mg/l cephalothin, 2 mg/l Fungizone, and 10% fetal bovine serum. Test media contained the same components as the plating medium except for serum source. Stromal-vascular cells were seeded in 25-cm2 flasks and 35-mm tissue culture dishes at a density of 1 x lo4 cells/cm2. Cells were cultured at 37°C in a humidified 5% CO, atmosphere. Plating medium was replaced with growth medium containing 2.5% porcine serum (Hyclone, Logan, UT) in medium 199 after 24 h of incubation. Growth medium was changed on days 1 and 3 of culture. Medium was changed to test media containing 2.5% test sera in medium 199 on day 5 of culture (confluency). Test media were changed every 3 days of culture. Cells were harvested on day 14 of culture and analyzed for activities of GPDH and LPL and for DNA content (15). Additional cells were plated in 25-cm2 flasks to evaluate glucose metabolism by the cells following differentiation in response to the test sera, according to previously published procedures (32). Briefly, cells were incubated with 2.5% porcine serum (Hyclone) in medium 199 from day 1 to day 5 of culture. Medium was changed to test media containing 2.5% test sera in medium 199 on day 5 of culture. Test media were changed every 3 days, and glucose metabolism was evaluated on day 14 of
Downloaded from www.physiology.org/journal/ajpendo at Washington Univ (128.252.067.066) on February 12, 2019.
SOMATOTROPIN
AND ADIPOSE
Cellswere washedfor 1 h with Earle’sbalancedsaltsto remove any medium contaminants and switched to incubation medium containing medium 199, 25 mM N-2-hydroxyethylpiperazine-N’ -2-ethanesulfonicacid (HEPES), 10 mM glucose,1 &i/ml [U-14C]glucose,and 2% bovine serumalbumin for 4 h. CO, was collected in plastic wells containing hyamine hydroxide, and the incorporated radioactivity was measuredby scintillation counting. Flask contents (porcine adipocytes) were analyzed for [ 14C]glucose incorporation into COe,total lipids, and fatty acids according to the procedure of DeCingolani (10). Isolation of total RNA. Tissue sampleswere pooled from animals in each treatment group. Total RNA was extracted and isolated using the guanidine isothiocyanate-cesium chloride method (6). Pooled tissues (5 g) were homogenized in 4 M guanidine isothiocyanate and centrifuged through a cushion of 5.7 M cesium chloride for 22 h at 20°C. RNA was phenolchloroform extracted and ethanol precipitated. Total RNA was quantified spectrophotometrically at 260 nm in a Gilford Responsespectrophotometer,then electrophoresedthrough a 1.2% denaturing agarosegel and stained with ethidium bromide to determine integrity of the RNA. All RNAs were judged to be intact by the presenceof 28s and 18s ribosomalbandsat a mass ratio of -2:l. Dot blot. Total RNA wasdenatured in 50% deionized formamide, 6% formaldehyde, and 20 mM Tris, pH 7.0. Various concentrations of total RNA were transferred to a charged nylon membrane (Immobilon, Millipore) using a dot-blot vacuum manifold (BRL, Gaithersburg, MD). Yeast tRNA was usedas the negative control and did not hybridize with the probes. Levels of mRNA on dot blots were estimated by laser densitometric scanning (LKB Ultrascan XL scanninglaserdensitometer) and by direct comparison to known concentrations of a cRNA standard curve present on the sameblot, according to procedurespreviously described(21). Northern analysis. Total RNA was isolated, and 20 pg were denaturedat 65°C for 5 min with 50% deionized formamide, 2.2 M formaldehyde, 24 mM HEPES, 6 mM sodiumacetate trihydrate, and 1.2 mM EDTA. Total RNA was separatedby electrophoresisin a 1.2% agarose-2.2M formaldehyde geland transferred to a nylon membrane(Immobilon, Millipore) by capillary elution transfer according to the procedure of Maniatis et al. (23). RNA filters were immobilized by baking in a vacuum oven for 2 h at 80°C. Ethidium bromide staining of ribosomalRNA was usedto monitor RNA integrity and equivalent loading of samples.Quantitative and equivalent transfer was verified by ethidium bromide staining and ultraviolet analysis of the gel following transfer. Nick trandation. Probeswere labeledby nick translation with specific activity of 1.5-2.0 x lo8 countsmin-l .pg DNA? Membranes for both dot-blot and Northern analysis were prehybridized in 5 ml of Rapid Hybridization Buffer (Amersham), 100 bg/ml yeast tRNA, and 1% Triton X-100 at 70°C for 30 min. Hybridization wasperformed at 65°C for 18 h. Membranes were rinsed at 65°C with 2~ SSC, 0.1% sodiumdodecyl sulfate (SDS) twice for 15 min, two times with IX SSC, 0.1% SDS, and twice in 0.1x WC, 0.1% SDS for 15 min each. All membranes wereautoradiographedusing Kodak XAR X-ray film at -70°C. Probes.The 580-bpporcine IGF-I cDNA wasa gift from Dr. Frank Simmen, and the rat 720-bp IGF-II cDNA was kindly provided by Dr. M. Rechler. The 720-bp rat IGF-I cDNA used for the cRNA transcripts was provided by Dr. P. Rotwein. Statistical analysis. Data were analyzed by two-way analysis of variance using SAS (35). The model includedthe main effects of diet, pST treatment, and the interaction of diet and treatment. Differences between means were tested using Fisher’s least significant difference when effects were significant (P < 0.05). culture.
E639
TISSUE
RESULTS
There was no significant effect of diet or somatotropin on rate of gain of animals in this study. The lack of effect of somatotropin on gain is in contrast to other studies (5, 13, 14) and can likely be accounted for by the small number of animals used in the present study. However, PST treatment did cause a reduction in feed consumption (-27%, P < 0.0001) and an improvement in feed conversion (feed-to-gain ratio: vehicle, 3.74; PST, 2.78; P < O.OOl), in agreement with previous work (5, 13, 14). Similarly, dietary fat resulted in a reduced feed intake (-8%; P < 0.05) and an improved feed-to-gain ratio (control, 3.67; fat 2.85; P c 0.0001). The effects of dietary fat were accounted for by differences in energy density of the diet. Thus, on a caloric basis, intake and feed conversion of animals fed fat were similar to those of animals fed the control diet. These results are similar to those reported previously (5). The effects of pST on serum growth hormone and IGF-I concentration are shown in Fig. 1. Treatment with exogenous pST elevated serum growth hormone and IGF-I concentrations in comparison with treatment with vehicle, independent of diet. Analysis of biopsies of subcutaneous adipose tissue which were utilized for tissue incubations and enzyme assays indicated that pST treatment and dietary fat affected the cellularity of the biopsies (Fig. 2). PST treatment resulted in more adipocytes per gram tissue than in tissue from control animals by determination of average cell number per gram tissue by osmium tetroxide fixation and electronic counting. Feeding of a high-fat diet resulted in fewer adipocytes per gram of tissue than in animals fed the control diet. Treatment with pST did not overcome the effect of fat feeding on the cellularity of the samples of subcutaneous adipose tissue. Treatment with exogenous pST resulted in an inhibition of glucose metabolism within the adipose tissue of treated swine when the results are expressed per lo6 cells (Fig. 3). Glucose conversion to carbon dioxide, total lipids, and fatty acids was reduced by at least 60%. Feeding of a diet supplemented with fat did not alter this pST response. Somatotropin treatment had minor inhibitory effects 2oflfl400 0
TRTl
TRT2
'
PST
0
~-1
1350
TRT3
Fig. 1. Effect of porcine somatotropin (PST) and dietary fat supplementation on serum PST and insulin-like growth factor I (IGF-I) concentration. Values represent mean of triplicate determinations of pooled samples as described in METHODS. TRT 1,O mg PST/day; TRT 2,2 mg PST/day; TRT 3,0 mg pST/day, 10% added fat; TRT 4,2 mgPST/day, 10% added fat.
Downloaded from www.physiology.org/journal/ajpendo at Washington Univ (128.252.067.066) on February 12, 2019.
E640
SOMATOTROPIN
AND ADIPOSE
1.4 Pooled n
4”
SE
0.62
g 3.5-’ .:.
-/
j= E cn
C. :s >: :ij: ::: j:i :t. j:::; :c: :;:.: j: :‘: j:::: ;$ ::: ;.i $ g :i: ... :.: j:.:. ::: ::: :.: ;g :i: ::: # :‘: :7 j: ;.: j: ;.: F. :_ 5. ;:I ;:.:. :::::
-/
yO.8-’ -/
;0.6-’ -E
-/
z’ 0.4-’ s -,
O-2-’ / 0’
8 “+
p ‘.. ._.
b
l-’
E
Pooled
C
I .2-’ 5 E
TISSUE
I
TRT 1
TRT 2
TRT 3
3- I
5 2.5-’ Q) ‘is E 2- Z
::: j::: ::: j:::: j:::: (..:. :.: :.: ::: :.: ::: :c :.. .:.
